How far do parenting programmes help change norms ...

Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=rgph20

Global Public HealthAn International Journal for Research, Policy and Practice

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/rgph20

How far do parenting programmes help changenorms underpinning violence against adolescents?Evidence from low and middle-income countries

Rachel Marcus , Jenny Rivett & Krista Kruja

To cite this article: Rachel Marcus , Jenny Rivett & Krista Kruja (2020): How far do parentingprogrammes help change norms underpinning violence against adolescents? Evidence from lowand middle-income countries, Global Public Health, DOI: 10.1080/17441692.2020.1776364

To link to this article: https://doi.org/10.1080/17441692.2020.1776364

Published online: 09 Jun 2020.

Submit your article to this journal

Article views: 129

View related articles

View Crossmark data

How far do parenting programmes help change normsunderpinning violence against adolescents? Evidence from lowand middle-income countriesRachel Marcusa, Jenny Rivettb and Krista Krujab

aOverseas Development Institute, London, United Kingdom of Great Britain and Northern Ireland; bIndependent

ABSTRACTRecent years have seen an upsurge in parent education programmes inlow and middle-income countries (LMICs) that aim to help reduceviolence against children. This article draws on a narrative review thatexamined the impact of 42 programmes working with parents ofadolescents in LMICs. Here we focus on 17 initiatives that aimed toreduce neglect of, or physical, emotional or sexual violence againstadolescents, or to reduce child marriage. Programmes aiming to preventsexual violence or child marriage generally focused more strongly onunderstanding and challenging prevailing norms, while those orientedto preventing physical and emotional violence emphasised sharinginformation and practising new communication skills. We argue that keyelements of programme design (group-based participatory sessions,formative research that enabled sensitive framing and adaptation ofcontent) have strong potential to help shift norms that underpinviolence against adolescents. To fulfil their potential to change normsunderpinning violence against adolescents, programmes should expandtheir reach, with a particular focus on embedding initiatives withininstitutions that can take them to scale, promoting male engagement,and support participants to maintain changes over the long-term.

ARTICLE HISTORYReceived 16 December 2019Accepted 30 April 2020

KEYWORDSParenting programmes;violence against adolescents;social norms; gender norms

Introduction

Recent estimates suggest that globally there is an ‘epidemic’1 of violence against children and ado-lescents. For example, data collected for the 2017 global report Ending Violence in Childhood(Know Violence in Childhood, 2017) suggest that 1.3 billion children aged 2–14 had experiencedcorporal punishment the previous year, and that 18 billion adolescent girls aged 15–19 had experi-enced sexual violence. Child marriage is increasingly considered a form of violence and affects anestimated 12 million girls a year.2

Social norms that ‘accept, support, or allow indifference to any forms of violence’ (Klika & Lin-kenbach, 2019, p. 2) are increasingly recognised as a structural factor underpinning violence againstchildren and adolescents (Lilleston et al., 2017; Know Violence in Childhood, 2017; World HealthOrganisation [WHO], 2018). Reflecting this, the INSPIRE package of interventions lays out seveninterconnected strategies for preventing and responding to violence against children, of which chan-ging norms and values is the second (WHO, 2018). Another key strategy outlined in the INSPIREpackage is parent and caregiver support. The extent to which parenting education programmescan be a vehicle for changing norms and values that underpin violence against children, thus con-tributing to two of these strategies, is the focus of this article.

© 2020 Informa UK Limited, trading as Taylor & Francis Group

CONTACT Rachel Marcus [email protected] 203 Blackfriars Road, London, SE1 8NJ

GLOBAL PUBLIC HEALTHhttps://doi.org/10.1080/17441692.2020.1776364

In recent years there has been an upsurge in parenting programmes aimed at helping parents andcaregivers3 in low and middle-income countries (LMICs) develop ‘positive parenting skills’. Parent-ing programmes can be defined as ‘activities oriented to improving how parents approach and carryout their role as parents and to increasing parents’ child-rearing resources, including, knowledge,skills and social support’ (this definition draws on Daly et al. (2015), p12). They are based on theassumption that improving parents’ understanding of adolescent development and strengtheningtheir communication skills will lead to better family relationships, improved care, less violence,and better mental health (of both parents and adolescents). A growing number aims to help preventsexual abuse and exploitation through raising parents’ and children’s awareness of risks and helpingthem develop skills to avoid risky situations.

Previous syntheses of the impacts of parenting programmes have focused exclusively on youngerchildren or have not disaggregated impacts on adolescents (eg Knerr et al., 2013; Wessels et al., 2013).Motivated by this gap, the authors recently conducted a narrative review of parenting programmes inLMICs that target the parents of adolescents (Marcus et al., 2019). In this article, we focus on a subsetof programmes that aimed to reduce neglect, physical, sexual or emotional violence against adoles-cents, or to contribute to reducing child marriage. We examine the impacts of these programmes,and how far they have been able to change norms that underpin violence against adolescents.

Conceptualising norms underpinning violence against adolescents and norm changeprocesses

Social norms involve shared beliefs about appropriate behaviour (sometimes termed ‘injunctivenorms’) and about typical behaviour (‘descriptive norms’). These norms are shared within a given‘reference group’ (community of people whose opinions matter to an individual) (Learning Colla-borative, 2017). The programmes we examined aimed – directly or indirectly – to address one ormore of three types of norms: those governing relationships between parents and children; normsrelated specifically to permissible violence; and gender norms, which intersect with other norms,as well as having their own specific effects. In many contexts, norms concerning parent–childrelationships emphasise parental authority and children’s obedience and involve limited discussion,and persist throughout adolescence (UNICEF, 2016). Norms related to permissible violence varyconsiderably, but include, in some contexts, the belief that physical punishment and/or harsh verbalcriticism can be necessary for raising children, at least in specific circumstances, such as deliberatemisbehaviour or certain kinds of mistakes (Doubt et al., 2018; Mejia et al., 2012; Sim et al., 2014) or toenforce norms of appropriate gendered behaviour (Basu et al., 2017). Within any given community,while individual attitudes vary, there is often a degree of consensus around the appropriateness ofthese norms. By contrast, the normative framework in most contexts overtly rejects child sexualabuse and forms of violence such as rape, but boundaries are often blurred around what constitutesconsent and what constitutes abuse or exploitation, particularly where transactional sex is common(Buller & Pichon, 2020).

Finally, gender norms that govern expectations of how male and female adolescents and adultsshould behave in different contexts, both exert their own influence on adolescents’ experiencesand on parenting, and intersect with the types of norms previously outlined. These are complex, con-text-specific and often nuanced. For example, norms around how parents discipline adolescents aregendered, with some evidence suggesting that there are stronger taboos on fathers hitting daughtersthan sons (Jejeebhoy et al., 2014). In many contexts, norms require female chastity but permit orencourage male (hetero-) sexual adventurousness, affecting both girls’ and women’s experiencewithin sexual relationships, their risk of sexual violence, and familial and community reactions ifthey are assaulted.

Interventions aiming to change norms may attempt to change either descriptive or injunctivenorms, or both, and may do so either explicitly and directly, or indirectly by addressing some ofthe factors that hold particular norms in place. To analyse the contribution and potential of

2 R. MARCUS ET AL.

parenting programmes to norm change, we adapt a framework developed by the Learning Collabora-tive (2017). We distinguish three main sets of strategies: changing attitudes through providing par-ticipants with new information and the opportunity to critically reflect on prevailing norms;changing behaviour through practising new skills such as positive discipline or communicationwith adolescents on formerly taboo topics, with the support of a peer group of other participants;and reaching a tipping point for change through organised scale-up and institutionalisation of pro-grammes. (See Table 1 which shows which programmes adopted elements of each strategy). In thefollowing sections, we examine evidence of programme effectiveness, and discuss how far the inter-ventions examined made use of each approach.

Methodology

This article is based on comprehensive searches of academic and grey literature, complemented withkey informant input. Database searches were undertaken in Web of Science, PsycINFO, Ovid, andEbscoHost in September 2018. Studies in English and Spanish which reported on the outcomes of anintervention involving the parents of adolescents, took place in an LMIC since 2000, and include acounterfactual were retained. Studies were excluded if their outcome focus was solely biomedical.Additional web searches were also carried out in December 2018 for relevant organisations, pro-grammes and authors, and studies meeting review criteria were snowballed from systematic reviewsand from included literature. Finally, targeted requests were made to key authors. (See Figure 1 for anoverview of the process). Fifty-eight studies of 42 programmes were ultimately included in the fullreview and provided insights on a range of adolescent wellbeing outcomes, including impacts on psy-chosocial wellbeing, substance abuse, and sexual and reproductive health knowledge. This articledraws on 27 evaluation studies4 that examine the impact of 17 programmes on violence againstadolescents.

Table 2 provides an overview of the programmes and studies discussed in this article. Themajority used rigorous quantitative designs (three randomised controlled trials and nine quasi-experimental studies); studies of eight programmes additionally reported some qualitative insights.5

All studies examined parents’ self-reported outcomes (such as knowledge, attitudes, behaviour orintended behaviour) and studies of 14 programmes also drew on adolescent reports for at leastone outcome.

Limitations

Only two6 studies disaggregated quantitative impacts by parent gender and only two studies did soby adolescents’ gender,7 limiting insights on gender differences in programmes’ impact. Most studiesonly reported on short-term effects: 10 endline evaluations took place within six months and anotherfour within a year of participation.

None of the quantitative data directly measures norms in the sense of people’s beliefs aboutexpected or typical behaviour in their communities. However, quantitative data on reported atti-tudes, behaviour and intended behaviour change give an indication of changes in the ‘bedrock’underpinning norms, and of the scale and direction of possible normative shifts. Qualitative dataindicated the processes by which changes had occurred, and limitations to these changes, and inthree programmes (Sinovuyo, CHAMP and Happy Families), provided more direct insights intonormative shifts.

Overview of programmes

Table 2 summarises key information about the programmes discussed in this review.The programmes examined took place in 23 countries, with the majority in Sub-Saharan Africa

(nine programmes) and Latin America (five programmes), and a roughly even spread between urban

GLOBAL PUBLIC HEALTH 3

Table 1. Norm change strategies used by programmes reviewed.

Norm changebuilding blocksand strategies

Change attitudes Practise new behaviours Reach a ‘tipping point’ for norm change

Provide accessto new

knowledge

Explicit discussionof social / gender

norms Practise new skills

Develop newdescriptive norm

throughperforming new

behaviour

Developsupportive

reference group tohelp sustainbehaviour

Action withstakeholders atmultiple levels

Increase reachthroughhorizontalscale-up

Institutionaliseprogrammes

Address broaderfactors thatunderpin

violence againstadolescents

Exampleprogrammes

Allprogrammes

Bihar PCCP;COMPASS;CHAMP; Choices-Voices-Promises;FMP;Happy FamiliesProgramme;Imbadu Ekhaya;Sinovuyo

CHAMP; FMP;Familias Fuertes;Happy FamiliesProgramme; GoGirls!;Imbadu Ekhaya;Let’s Talk;Sinovuyo

Not documentedin any evaluation

Sinovuyo;CHAMPHappy Familiesprogramme

No evidence FMP;Sinovuyo

No evidence CHAMP;School forParents;Sinovuyo

4R.M

ARC

USET

AL.

and rural or peri-urban locations. Nine programmes worked only with parents and with eight withboth parents and children, either separately, or together for joint learning and practising skills.8 Chil-dren of participating adults spanned the adolescent age range, with ages 11–14 most frequently rep-resented. In the vast majority of programmes (16/18) adult participants were mostly or entirelyfemale, and in only one (Let’s Talk in South Africa) were they were mostly male. Most programmestargeted the parents of both boys and girls; two – Go Girls! and COMPASS – only included girls andtheir parents.

Almost all programmes (16/18) were delivered through group classes, held in settings such ascommunity centres and schools, usually weekly. One initiative (Parceria Project) was delivered viahome visits, and one (Ligue 132) via telephone-based sessions. Programmes varied in length,from two to 16 sessions, with the majority involving five to seven sessions. Three programmeswere implemented over the course of a year; the others spanned several weeks to months.

The majority of programmes were small-scale, often pilot projects.9 Eight programmes reachedunder 100 parents, five programmes worked with 100–500 participants (parents and children com-bined) and four reached over 500 participants. Three of these (FMP, Strengthening Families Pro-gramme/ Familias Fuertes and Sinovuyo) were national iterations of large multi-countryprogrammes, reaching considerably more families than indicated by national evaluations. Forexample, FMP has cumulatively reached over one million participants to date (K. Miller, personalcommunication, April 30, 2019).

Programme impacts on norms about different forms of violence

Table 3 summarises evidence of programmes’ impact on attitudes and behaviour, the main indi-cators of norm change in these studies.

Physical and emotional violence against adolescents

Most programmes took place in settings characterised by acute poverty, high levels of communityviolence, conflict, displacement, HIV-related illness and death, and domestic violence. In such con-texts, some parents reported viewing violence as a protective strategy to ensure obedience and reduceexposure to risk. For example, participants in the Strengthening Families Programme in Panama,who lived with daily realities of severe poverty, and drug-related violence, indicated that harsh par-enting styles were the norm:

We will not have the softness to tell a boy ‘come here and do this’. We are going to be tough and yell (Mejiaet al., 2012, p. 61)

In the Happy Families Programme working with Burmese migrant families in Thailand, manyparticipants cited learning how to teach their children to be obedient as their motivation for

Figure 1. PRISMA diagram.

GLOBAL PUBLIC HEALTH 5

Table 2. Overview of programmes.

Programme Area of focus

Country andgeographicallocation Approach

Programmeduration(parent

component) Who participated

Age ofadol-escents

Evaluation design/methods in studies

reviewed

Programme Scale/ Samplesize, as described in each

studya

Bihar parent–child pilotcommunicationprogramme (BCCP)(Jejeebhoy et al., 2014)

Violence and abuseprevention, genderequity, familyrelationships andcommunication, SRH

IndiaRural

Group classes incommunitysetting

16 sessions,5–9 months

Parents (moremothers thanfathers) andadolescentstogether andseparately

13–17 Pre-post (nocontrol group)

478 mothers, 357 fathers,362 daughters, 429 sons

Collaborative HIVPrevention andAdolescent MentalHealth FamilyProgramme(CHAMP) –Amaqhawe(Bhana et al., 2004)(pilot)(Paruk et al., 2009); (Bellet al., 2008)

Family relationshipsand communication,SRH

South AfricaRural,peri-urban

Group classes incommunitysettings

10 weeks/sessions

Parents (moremothers thanfathers) only

9–13 Quasi-experimental(Bhana et al.,2004; Bell et al.,2008); qualitative(Paruk et al.)

Main programme: 478caregivers, 557 children(Bell et al. 2008), Pilot: 124families (72 intervention,52 control) (Bhana et al.);Qualitative study: 9women from pilot area(Paruk et al., 2009)

Choices-Voices-Promises(Lundgren et al., 2018)

Gender equity NepalRural

Group classes incommunitysettings andschools

2 sessions, &once weeklyfor 3 months

Parents (moremothers thanfathers) andadolescentsseparately

10–15 Quasi-experimental;qualitativecomponent

1200 adolescents (600intervention, 600 control);600 parents (300intervention, 300 incontrol)

Creating Opportunitiesthrough Mentorship,ParentalInvolvement, andSafe Spaces(COMPASS)(Stark et al., 2018)

Violence and abuseprevention, genderequity, familyrelationships andcommunication

DemocraticRepublic ofCongon/a

Group classes 13 sessions,1yr

Parents (moremothers thanfathers) andadolescentsseparately

10–14 RCT 764 caregivers (389intervention arm, 375control); 869 adolescentgirls (426 intervention, 423control).

Familias Fuertes(Corea et al., 2012);(Orpinas et al., 2014);(Vasquez et al., 2010);(PAHO, 2006)

Family relationshipsand communication,substance abuse; in ElSalvador also: SRH,violence and abuseprevention

Bolivia, Chile,Colombia,Ecuador, ElSalvador,HondurasUrban,peri-urban

Group classes inschools

7 weeks/sessions

Parents (moremothers thanfathers) andchildrentogether andseparately

10–14 Quasi-experimental;qualitative datafrom El Salvador

119 families in Bolivia, 182families in Colombia, 82families in Ecuador(Orpinas et al., 2014); 41parents and theiradolescents in Honduras(21 parent-adolescentpairs intervention, 20

6R.M

ARC

USET

AL.

control) (Vasquez et al.,2010); 76 parents andcaregivers in El Salvador(40 intervention, 36control), 86 adolescents(46 intervention, 40control) (PAHO, 2006); InChile 124 parents and 149adolescents intervention,223 adolescents and 165parents control (Coreaet al., 2012)

Families Matter!(Vandenhoudt et al.,2010); (Kamala et al.,2017)

SRH; violence andabuse prevention

Kenya, TanzaniaRuralZimbabwe,Urban

Group classes incommunitysettings

5–7 sessions Parents (moremothers thanfathers) andchildrentogether andseparately

9–12 Quasi-experimental

658 parents and children(parent-child pairs)participated in Tanzania(Kamala et al., 2017), 375parent-child pairs in Kenya(Vandenhoudt et al., 2010)and 248 parent-child pairsin Zimbabwe (Shaw et al.,2019)

Go Girls! Initiative(Schwandt &Underwood, 2013)

Family relationshipsand communication;SRH; violence andabuse prevention

Botswana,Malawi,MozambiqueUrban, Rural

Group classes incommunitysettings

11 sessions,1yr

Parents (moremothers thanfathers) only

11–18 Quasi-experimental;qualitativecomponent

Sample sizes by country: 401(299 intervention) inBotswana, 414 (386intervention) in Malawi,and 603 (555 intervention)in Mozambique. Controlswere those adults who didnot participate.

Happy Familiesprogramme(Annan et al., 2017);(Puffer et al., 2017);(Sim et al., 2014)

Family relationshipsand communication,mental health/ well-being promotion

ThailandUrban, Rural,Peri-urban

Group classes incommunitysettings

12 weeks/sessions

Parents (moremothers thanfathers) andchildrentogether andseparately

7–15 RCT; qualitativecomponent

479 children and 513caregivers (256 caregiversintervention, 257 control)

Imbadu Ekhaya(Parents Matter!)(Armistead et al., 2014)

Violence and abuseprevention, genderequity, SRH

South AfricaUrban

Group classes,unclear setting

6 weeks/sessions

Parents (allmothers) andchildrentogether andseparately

10–14 Quasi-experimental

90 female parents and theiryouth with a birthdayclosest to theinformational meeting (52intervention, 38 control)

(Continued )

GLO

BALPU

BLICHEA

LTH7

Table 2. Continued.

Programme Area of focus

Country andgeographicallocation Approach

Programmeduration(parent

component) Who participated

Age ofadol-escents

Evaluation design/methods in studies

reviewed

Programme Scale/ Samplesize, as described in each

studya

Let’s Talk(Bogart et al., 2013)

SRH South AfricaUrban

Group classes incommunitysettings

5 weeks/sessions

Parents only(more fathersthan mothers)

11–15 Quasi-experimental

66 parents, 64 adolescents

Ligue 132(Valente et al., 2018)

Substance abuse Braziln/a

Telephonesessions

4 sessions, 1month

Parents (moremothers thanfathers) only

10–18 Quasi-experimental

26 parents initially – finaldata used for analysisbased on 15 parents (9intervention, 6 control)

Parceria project(Pereira et al., 2013)

Violence and abuseprevention

Braziln/a

Home visits 16 weeks/sessions

Parents (allmothers) only

12–16 Pre-post (nocontrol group)

17 mothers (all completedintervention)

Parenting for LifelongHealth: SinovuyoTeen(Cluver et al., 2016);(Doubt et al., 2017);(Doubt et al., 2018);(Cluver et al., 2018)

Violence and abuseprevention, familyrelationships andcommunication

South AfricaRural,peri-urban

Group classes incommunitysettings,home visits, andfinancialplanningworkshops

14 weeks/sessions(Pilot: 7sessions)

Parents (moremothers thanfathers) andchildrentogether andseparately

10–18 RCT (main phase),Quasi-experimental(pilot);qualitativecomponent

Pilot: 119 adolescents and119 caregivers (Cluveret al., 2016); Mainprogramme: 552 families(270 intervention; 282control) (Cluver et al.,2018); 240 participants infocus group discussionsand 42 participants ininterviews, also 9workshop observationsand 280 facilitator notesfrom sessions (Doubtet al., 2017 & 2018)

ParentingpsychoeducationinterventionJordans et al. (2013)

Violence and abuseprevention, familyrelationships andcommunication,mental health/well-being promotion

BurundiRural

Group classes inschools

2 sessions Parents (moremothers thanfathers) only

10–14 Quasi-experimental;qualitativecomponent

58 children and theirparents in intervention, 62children and their parentsin control group

School for ParentsprogrammePereira Lima et al.(2007)

Family relationshipsand communication,empowerment

BrazilUrban

Group classes,individualisedsupport forparents, settingunclear

9 sessions, 1year

Parents (moremothers thanfathers) only

Childrenunder18

Pre-post (nocontrol group);qualitativecomponent

48 parents (all completedintervention)

SRH education forparents ofadolescents withintellectual

SRH Turkeyn/a

Group classes incommunitysettings

16 sessions, 5–9 months

Parents only(equal gendersplit)

10–19(61.9%aged10-14)

Pre-post (nocontrol group)

42 mothers and 42 fathers(no control)

8R.M

ARC

USET

AL.

disabilitiesKok and Akyüz (2015)

Strengthening FamiliesProgramme: ForParents and Youth10–14 (SFP 10–14)Maalouf and Campello(2014)

Family relationshipsand communication,substance abuse,violence and abuseprevention

Guatemala,Honduras,Panama,SerbiaUrban

Group classes inschools

10 weeks/sessions

Parents (moremothers thanfathers);Unclear ifadolescentsparticipated

10–14 Pre-post (nocontrol group)

240 children and 218parents in total (nocontrol):40 children and 40 parentsin Serbia, 83 children and78 parents in Panama, 44children and 42 parents inGuatemala, 73 childrenand 58 parents inHonduras

aBecause all the initiatives examined were run on a pilot basis, the sample size for the evaluation was identical to the total number who participated (excluding loss to follow-up), or further informationabout programme scale was not given.

GLO

BALPU

BLICHEA

LTH9

Table 3. Summary of programme impacts on violence against adolescents.

ProgrammePhysical and Emotional

ViolenceSexual violence (SV) and child

marriage (CM) Neglect

Bihar PCCP Parent-reported changes inperpetration of violence:Mothers vs sons: 29%decrease; fathers vs sons17% decreases; Mothers vsdaughters 12% decrease;fathers vs daughters: 5%Adolescent-reportedchanges in violenceperpetration:Mothers vs sons 28%decrease; fathers vs sons36% decrease; mothers vsdaughters 3% decrease;fathers vs daughters 10%decrease

Increase in mothers’acceptance of children’schoice in who they marry(15% for boys and 25% forgirls). 7% decrease infathers’ acceptance thatboys should have a choice inwho they marry; 13%increase in acceptance ofgirls’ right to choice.Increase in mothers’acceptance of children’schoice of when they marry(7% for boys and 16% forgirls). Respective increasesamong fathers: 7% and 6%.

CHAMP Parent-reported reduction inuse of harsh physicalpunishment (qualitativeevidence).

More collective monitoring andguidance of children

Choices-Voices-Promises Average increase of 7 monthsin age parents desire theirdaughters to marry (19.4 –20); just 1% reduction (notstatistically significant) inproportion of parents whoreported considering thatCM is ‘bad for thecommunity’.

COMPASS No reduction in parental useof physical punishment; noimpact on girls’ likelihood ofexperiencing all forms ofviolence.

Greater decrease in controlgroup reports ofexperiencing sexual violencethan intervention group (C –26%; I – 23%); coerced sex(C – 27%; I – 15%); and childmarriage (C – 14%; I – 5%).

Both the control andintervention group reported a10% decrease in feelinguncared for; reported increasein perceived parental warmthand affection towards theirchildren.

Familias Fuertes Greater understanding ofharm that physicalpunishment can cause; 88%increase in parent reports of‘never’ shouting atadolescents when angry,54% decrease in reports of‘often’, and 31% decrease inreports of ‘always’ shoutingat adolescents when angry(Chile). Statisticallysignificant reduction inparent-reported ‘parentalhostility’ including shoutingand hitting (Bolivia,Colombia, Ecuador). Parent-reported reduction inphysical punishment ofadolescents (Qualitativeevidence, El Salvador).

Non-significant increase inparental monitoring ofadolescents (Honduras).Statistically significantincrease in parent-reportedparental monitoring(Ecuador).

Families Matter! Significant increase of parentalmonitoring of SV; increasedcommunity-level discussionof child sexual abuse andparental knowledge ofsupport services(Zimbabwe).

61% of parents and 62% ofchildren (Kenya) and 76% ofparents and 74% of children(Tanzania) reported increasedpositive parental monitoring.

(Continued )

10 R. MARCUS ET AL.

Table 3. Continued.

ProgrammePhysical and Emotional

ViolenceSexual violence (SV) and child

marriage (CM) Neglect

Go Girls! Initiative Parent-reported reduction inbeating and yelling(qualitative evidence).

Happy Familiesprogramme

Parent-reported 18%reduction in beating, yellingand swearing at children;adolescent-reported 15%reduction in beating.

Imbadu Ekhaya Increased parent-adolescentcommunication on sexualcoercion but no change inbroader gender normsunderlying SV. Effectsreduced at 6-month followup – though they remainedin parent reports theyalmost disappeared inadolescent reports.

Let’s Talk Statistically significantincrease in participantparents discussing sexualcoercion with adolescents.

Ligue 132 Statistically significantreduction in parent-reportedphysical abuse againstadolescents.

Statistically significantreduction in parent-reportedneglect and increase inpositive parental monitoringof adolescents.

Parceria project No impact on risk of parentalabuse but improved familycommunication.

Statistically significantimprovement in parent – andadolescent-reported parentalstyles including parentalmonitoring of adolescents.

Parentingpsychoeducationintervention

Parent-reported reduction inphysical punishment(qualitative evidence).

School for Parentsprogramme

Average of 78.5% increase inschool enrolment of under-18s at follow-up.

SFP 10–14 Statistically significantincrease in parent- andadolescent-reported calmcommunication anddiscipline without shoutingby parents.

Sinovuyo Teen Parent-reported reduction inphysical and emotionalabuse at 1-month and 5–9months post-intervention inpilot programme.Adolescents reported areduction in abuse at 1-month but no change at 5–9months. Parent-reportedreduction in use of corporalpunishment, but no changein adolescent reports at 1-month post-intervention(main programme).

No statistically significantimpacts on adolescentexposure to sexual abuse,however larger sampleneeded for an accuratemeasurement.

Reduced parent ‘poorsupervision’ and increased‘positive monitoring’ found inpilot but not mainprogramme; increasedparent-child communication.

SRH educationprogramme for parentsof adolescents withintellectual disabilities

No statistically significantimpacts on parents’ self-efficacy in protectingchildren from sexual abuse.

GLOBAL PUBLIC HEALTH 11

participating and reported relatively frequent use of harsh verbal punishment such as scaring theirchildren, swearing, and shouting, (Sim et al., 2014), though at baseline they also reported ‘rarely’taking out anger on their children (Puffer et al., 2017, p. 11).

Eleven programmes aimed to reduce physical and/or emotional violence10 against adolescents,through improving family communication and general parent–child relationships, and by helpingparents develop ‘positive discipline’ skills; the evaluations reviewed show evidence of a reductionin nine of these (Table 3). Studies of two programmes (COMPASS in DRC and Parceria Projectin Brazil) found no evidence of reduced violence. Two – the studies of Sinovuyo Teen and HappyFamilies – found mixed evidence with reductions on some indicators but not others. The scale ofchange (where reported) varied considerably between programmes (Table 3).

These programmes used two main strategies from the framework outlined in Table 1: helpingparticipants develop new knowledge and attitudes, and enabling parents to develop new modes ofbehaviour, which over time could become normalised in their families and reference groups.

Changing attitudes through new knowledgeQualitative data from South Africa indicates that programmes affected parents’ attitudes towardsharsh discipline by raising awareness of the physical and psychological harm it causes adolescentsand its ineffectiveness:

We have learnt that we were abusing our authority over our children. We learnt that the treatment we gave ourchildren sometimes had bad results. (Paruk et al., 2009, p. 64)

If you raise your voice to a child or beat her, she will completely ignore you. I was like that before … (Doubtet al., 2018, p. 25)

Participants also reported learning that refraining from using violent discipline did not mean thatthey could no longer carry out their responsibility to educate their children:

I learnt that there is not [a need] to shout to a child in order to get your point across. I should be calm, sit himand gather the facts. So that he could be at ease to tell me. I should not raise my voice at him and beat him.However, I must show him that I am disappointed in what he did. (Doubt et al., 2018, p. 25)

Parents participating in CHAMP in South Africa explained that the programme had helped themunderstand their rights in a context where they believed children’s rights meant they were unable todiscipline them:

We couldn’t talk to our children the way we wanted…when you were instructing the child or smacking thechild for something that she/he had done wrong, the child would tell you that she/he is going to take you tocourt… . You wouldn’t feel like a parent to the child, but felt that the child was more powerful than you…CHAMP was able to solve that problem by teaching us parents how far children’s rights go and how far parents’rights go…We were able to have a proper discussion with our children and there was good communication…and the child was able to realise that she/he is still a child and this is a parent. (Paruk et al., 2009, pp. 63–64)

Developing new norms through practising new behaviourHelping parents communicate more effectively with their children, through developing new normsaround parent–child communication, and by practising communication skills through role-playswas the most common programme approach to reducing parent violence against children. Aswell as practising talking with adolescents, adult Sinuvoyo Teen participants reported learningtips for managing anger and stress:

I learned that I should not take out stress on my child, I… take a pause or do anything like read a book todistract it – then the stress would disappear. (Loening-Voysey et al., 2018, p. 26)

Adolescent Sinovuyo participants reported adopting similar approaches when in disagreementwith siblings, and as a result, perpetrating less physical violence against them (Doubt et al., 2018).

12 R. MARCUS ET AL.

The evaluation of a parenting psychoeducation initiative in Burundi also recorded some reduction inadolescent violence as a result of programme participation (Jordans et al., 2013).

The studies reviewed found both quantitative and qualitative evidence of behaviour change. Forexample, Maalouf and Campello’s (2014) study of the Strengthening Families Programme in Gua-temala, Honduras, Panama and Serbia found that the programme led to a statistically significantincrease in parents’ self-reported agreement with the statement: ‘I tell my child when I am irritatedwithout blaming or criticising’. A mother participating in Go Girls in Botswana commented:

This program has helped me build a better relationship with my children. I used to be very strict with my chil-dren – I would beat them or yell at them. I attended the Go Girls! Adult–Child Communication program andnow I can sit with them and tell them that I love them. I can give them advice without beating them or yelling atthem. (Schwandt & Underwood, 2013, p. 1182)

As participants in Sinovuyo Teen in South Africa, reported, these changes in communication rep-resented a substantial shift in norms:

As black people we… do not want to speak to our children about crucial matters. They taught us to commu-nicate with our children, spend time with them and not sideline. them on issues. (Doubt et al., 2018, p. 22)

Parent neglect of adolescents

Encouraging parents to implement newly learnt communication skills was also the key strategy usedby programmes aiming to reduce neglect of adolescents, examined in six studies. These studies wereprimarily quantitative, with limited qualitative insights on processes of norm change around engagedparenting. The evaluations that drew on parents’ reports found positive impacts on indicators suchas increased engagement with children, increased positive monitoring of adolescents (Orpinas et al.,2014; Valente et al., 2018), and in the case of School for Parents, school enrolment (Pereira Limaet al., 2007).

The studies that drew on adolescent reports found more mixed results. For example, Stark et al.’s(2018) study of the COMPASS programme found no change in adolescent girls’ reports of feelinguncared for, though it did find an increase in parental warmth and affection towards their children.The evaluation of Sinovuyo Teen’s pilot programme found reports of reduced parent ‘poor super-vision’ and increased ‘positive monitoring’ from both adolescents and parents, while that of the sub-sequent larger-scale programme found no impacts on the same indicators (Cluver et al., 2016; Cluveret al., 2018). The timing of measurement (2–6 weeks after the intervention in the pilot evaluation, 5–9 months post-intervention in the main programme) may explain these findings as impacts mayhave started to fade over time.

Qualitative research from Sinovuyo Teen suggests that the changes reported arose from parentsand adolescents spending quality time together both during, and as a result of participation in, theintervention (Doubt et al., 2017, p. 773). This led both to increased communication and greatermutual respect. For example:

We share our problems. And that makes us close. (Teen, Doubt et al., 2018, p. 22)

We sit down and talk and it is really nice. He tells me about what goes on at school and he has really pushedhimself. He even plays cricket, they received a trophy and I would praise him. (Caregiver, Doubt et al., 2018,p. 22)

These changes are early indications of shifts in behaviour that may, over time, start to form newnorms around parent–adolescent interaction, and parental engagement with adolescents. Throughstrengthening bonds between participants, some of the group programmes (such as CHAMP andSinovuyo Teen in South Africa) also strengthened community social networks and helped reinstatenorms about community responsibility for supporting and guiding children and adolescents that hadweakened over time. As a CHAMP participant observed:

GLOBAL PUBLIC HEALTH 13

You find a child [who is not in school] and you send or accompany her/him to school…And then the mothercomes to you and says – I heard that you sent my child to school. Thank you very much, that was very helpful.Before, the mother would have said –What is the matter with you? That was not your business.…Now, there isthat spirit of togetherness, that I have seen the child doing wrong, and let me correct her/ him. (Paruk et al.,2009, p. 65)

None of the evaluations of initiatives aiming to reduce physical or emotional violence or neglectcomment directly on whether programmes stimulated discussion among participant parents aboutprevailing norms around harsh punishment, or supervision of adolescents. The evidence outlinedabove suggests a stronger focus on changing individual beliefs and behaviour, and on changethrough creating a new descriptive norm, rather than through participants deliberating on prevailingnorms. This contrasts with the programmes aiming to reduce sexual abuse and child marriage dis-cussed in the next section, which had a stronger focus on challenging norms as well as practices.

Sexual violence and child marriage

Studies of five programmes (COMPASS, FMP, Let’s Talk and Imbadu Ekhaya and an un-namedSRH education programme in Turkey) explored their effectiveness in helping parents protect ado-lescents from sexual violence, while evaluations of three programmes (Bihar PCCP, Choices-Voices-Promises and COMPASS) examined their impact on support for or incidence of child mar-riage. Five of these seven studies11 recorded some degree of positive change in terms of greatercapacity to protect adolescents from sexual violence, and reduced support for child marriage. Incommon with the programmes discussed in the previous sections, these initiatives provided infor-mation and enabled parents to practise new communication skills. All also involved some criticalreflection on prevailing gender norms.

Part of the curriculum for Let’s Talk, a worksite-based SRH and parenting education programme,and Imbadu Ekhaya, a similar, community-based initiative, focused on recognising violence andabuse in relationships and avoiding sexual coercion. By naming violence within relationships andmaking it visible, these programmes were helping to challenge norms that condone such violenceand consider it beyond challenge. The evaluation of Let’s Talk found a statistically significantincrease in participant parents reporting discussing sexual coercion with their adolescent childrenover the course of the programme. By contrast, among the control group, reported discussion of sex-ual coercion slightly decreased (Bogart et al., 2013). The evaluation of Imbadu Ekhaya included dis-cussion of sexual coercion and consent in its measure of ‘breadth of communication’ betweenparents and children and found statistically significant evidence of change on these measures. How-ever, it also found little shift in broader gender norms which underpin vulnerability to sexual vio-lence and abuse, which the evaluators attribute to the relatively limited time (one session out ofsix) devoted to discussion of gender norms (Armistead et al., 2014). In a similar vein, Starket al.’s (2018) evaluation of the COMPASS programme in the DRC found that it had had noeffect on adolescent girls’ exposure to any form of sexual violence or incidence of child marriage.The evaluation attributed this lack of impact to the crisis context (for which the curriculum was per-ceived to lack relevance), the infrequent parent meetings (once a month), which were insufficient toraise awareness and build commitment to act, and to the focus on girls and their parents, rather thanworking to change norms in the wider community, and in particular, among potential perpetratorsof sexual violence or abuse.

The Families Matter! programme (FMP)12 provides an example of a parenting initiative which,through a strong focus on understanding and challenging norms, has proven effective in enhancingparents’ understanding of sexual abuse and exploitation, and strengthening their capacity to protecttheir children. The module on child sexual abuse helps parents understand the gendered norms theirchildren may be having to negotiate, including norms that encourage being sexually active, and thesocial ridicule that girls can experience if they lack material goods that others have, such as mobilephones (CDC, 2019). Unusually among the programmes we examined, Families Matter encourages

14 R. MARCUS ET AL.

discussion of ‘pressures to conform to norms of masculinity, which may include alcohol and drugs inthe context of male group socialising, pressure to be sexually aggressive or else suffer social exclusion,and embarrassment to seek out information about sex’ (Miller et al., 2016, pp. 414–415).

Like many of the programmes discussed in this article, FMP sessions include role playing to helpparents and adolescents recognise, and practise skills for negotiating potentially risky situations, suchas:

What would you do if… you are walking through the neighbourhood when the tailor calls you over and whis-pers in your ear that he’s made you a beautiful dress: why don’t you come into his workshop and try it on?

What would you do if… the neighbour who has been kindly paying your school fees since your father lost hisjob asks you to come over to his house that evening? (Miller et al., 2016, pp. 414–415)

FMP also makes use of audio-narratives modelling good parent–child communication, includingencouraging parents to discuss consent with both sons and daughters (Miller et al., 2016). Researchin urban Zimbabwe found that after taking part in sessions based on this module, parents and chil-dren reported significantly higher levels of parental monitoring about child sexual abuse. Signifi-cantly more parents also reported conversations with people in their community about childsexual abuse and knowledge of where to access services if their child was abused (Shaw et al., 2019).

Two studies examined changes in attitudes and norms around child marriage. Jejeebhoy et al.’s(2014) study of the Bihar parent–child communication intervention in India found a statistically sig-nificant change in mothers’ acceptance of sons and daughters having a say in their marriage partners(but no change for fathers). The study of Choices-Voices-Promises in Nepal found that taking part inworkshops about raising adolescents in a gender-equitable manner did not lead to increased agree-ment among parents that ‘it is bad for a community if girls marry late’, and indeed that parents whohad taken part were less likely to disagree with this statement (i.e. were more likely to support earlymarriage), though it had a small positive effect in raising the average preferred age of marriage.Parents also reported a strong sense of social pressure to marry their daughters by their early 20s,to avoid gossip that might harm family reputations (Lundgren et al., 2018). Unsurprisingly, thiswas unchanged through a few community education sessions.

Strategies for strengthening programmes’ impact

Overall the evidence discussed in the previous sections indicates that parenting programmes can beeffective at reducing physical and emotional violence and neglect of adolescents, increasing parents’capacity to protect children from sexual violence and reducing their support for child marriage. Theevidence presented above also shows the limitations of relatively short programmes in changingingrained patterns of behaviour upheld by prevailing social norms – with changes on some indi-cators and not on others, and sometimes contradictory reports from parents and adolescents. Inthis section we discuss how programmes’ impact may be strengthened. These recommendationsare organised around two themes: widening programmes’ reach and increasing sustainable impacts;and issues related to programme structure and content.

Widening programme reach and increasing sustainable impacts

Changing norms that underpin violence against adolescents requires a critical mass of people to beexposed to new ideas and practices. With the exceptions of FMP, Familias Fuertes/ SFP and Sino-vuyo, which is currently being scaled up and replicated (Cluver et al., 2017), most of the programmesexamined were small pilot initiatives with the majority reaching fewer than 500 families. There issome evidence of programmes’ reach extending beyond immediate participants, through familymembers reading printed handouts and materials and discussing programme learning, neighbourstaking part in home visits alongside people who were registered for programmes, and some

GLOBAL PUBLIC HEALTH 15

spontaneous diffusion activities, such as Sinovuyo participants sharing their learning through churchgroups (Cluver et al., 2018). However, overall, the numbers reached remain low. For parenting pro-grammes to make a significant contribution to changing norms that underpin violence against ado-lescents, they would need to be implemented on a substantially larger scale (scaled up horizontally)(Carter et al., 2019). As the INSPIRE handbook shows, such horizontal scaling can be done cost-effectively and, where programmes are effective, can help prevent ‘downstream’ costs to individualsand society (WHO, 2018). In the next section, we discuss ways that this scaling-up might beachieved. It should be noted that these recommendations emerge from analysis of successes andgaps across the set of programmes we examined, rather than from quantitative studies that havedemonstrated the efficacy of specific approaches.

Institutionalising programmes. Many programmes were piloted by researchers and/or inter-national donors in collaboration with NGOs or, less frequently, government departments. Noneof the studies examined discussed how initiatives were, or could be embedded within public services,such as public health programmes, or offered through schools; one study (of a SRH-focused initiativein the wider review) piloted a model of working through churches. These approaches deserve furtherconsideration as they represent a possible way of extending programmes’ reach and longevity, andthus the likelihood of sustainable norm change (Learning Collaborative, 2019). They also provide ameans of engaging ‘norms influencers’ – people whose opinions and behaviour exert particular influ-ence on others, often by virtue of their social position and networks. Our wider review found thatthese included community health workers and priests (Jejeebhoy et al., 2014; Puffer et al., 2016).

Investing in strengthening social networks among participants. Strengthening social support net-works among programme participants and former participants may be a promising way to sustainprogramme impacts and increase their chance of contributing to changing norms among partici-pants, and in the wider community. Community-building activities, such as eating together, songsor prayers can help strengthen these bonds (Doubt et al., 2018; Sim et al., 2014). Sinovuyo Teenalso entrusted programme volunteers as ‘buddies’ who offered peer support between sessions.These were particularly important for participants with low levels of literacy, who struggled toread printed materials, but also helped reinforce the sense of a community among participantsmore broadly (Cluver et al., 2016).

As a CHAMP participant in South Africa reported:

with people that attended the programme, friendship and trust did develop. Since we met, we bonded so muchthat it came to a point where when you have a problem… you go to your friend that you met when youattended the programme. We are now able to help each other and phone each other as neighbours. (Paruket al., 2009, p. 65)

At least one Sinovuyo group resolved to continue meeting after the end of the parenting course(Doubt et al., 2018). As well as acting as a personal and social support source, informal social net-works of this kind remind parents that they are part of a reference network of people in their com-munities who are trying to parent in a different way.

Reaching a wider range of participants, particularly menConsistent with other reviews of parenting programmes (such as Bacchus et al., 2017; Panter-Bricket al., 2014), the main participants in all but one of the programmes we examined were women. Amajor reason for this was gendered norms ascribing primary responsibility for parenting to women,and responsibility for breadwinning to men. This meant that attending parenting programmes wasgenerally considered women’s responsibility, particularly where programmes did not also offer econ-omic strengthening activities and/ or where parents were not living together. In some cases, sessionswere held at times that clashed with men’s work commitments. As a consequence of these genderedpatterns of attendance, some women participants in South Africa and El Salvador commented thatthey were constrained in implementing their new learning because their male partners had not taken

16 R. MARCUS ET AL.

part and continued to parent in authoritarian ways (Loening-Voysey et al., 2018; Pan-AmericanHealth Organisation, 2006).

The single exception where men comprised the majority of participants – Let’s Talk in SouthAfrica – ran at workplaces, with parents being given time off work to attend (Bogart et al., 2013).This potentially promising strategy (well established in HIV prevention but less so in parenting edu-cation) deserves further exploration, as does partnering with religious institutions (Marcus et al.,2019). The growing number of programmes engaging fathers with the care of young children,such as Programme P (ABAAD & Promundo, 2019; Doyle et al., 2018) and REAL Fathers Initiative(Bacchus et al., 2017) may also have lessons for initiatives working with the fathers of adolescents.

Programme structure and content

As well as scaling up, our review suggests that effectiveness of parenting programmes can beincreased through changes to programme structure and content to increase the depth of attentionto norms that underpin violence against adolescents.

Increasing programme lengthMany of the programmes we examined involved relatively few sessions, five to seven being mostcommon. Some evaluations (e.g. COMPASS, Imbadu Ekhaya) indicated that programme lengthor frequency of sessions was insufficient to change attitudes, norms and behaviour. There is atrade-off between programme length and attractiveness to participants; while longer courses mayenable stronger bonds between participants to develop and can facilitate greater exploration ofdeep-seated norms, they can only do so if participants are not put off by programme length, dropout or attend irregularly. Longer programmes, such as Sinovuyo managed to retain participantsover a three-month period, but attendance levels were moderate (around 60%). Irregular attendanceled to Sinovuyo also offering ‘catch up’ home visits, which participants appreciated, but were morecostly, more demanding of facilitators, and therefore presented trade-offs for scalability (Loening-Voysey et al., 2018). A compromise may be to add one or two sessions – for example, FMP hasrecently added an additional session to strengthen learning on sexual abuse, which as discussedabove, has had substantial positive effects on parents’ knowledge and preparedness to prevent sexualabuse.

Linking parenting education to other poverty reduction and developmental activitiesRecognising that poverty-related stresses contribute to and exacerbate violence against adolescents,three programmes in our wider review provided parent–adolescent workshops on financial man-agement and budgeting. Qualitative evidence suggests that both parents and adolescents foundthese workshops beneficial, and, in the case of Sinovuyo, to contribute to improved economic well-being among participant families (Cluver et al., 2018). Inclusion of meals aids concentration andbuilds social connections among participants, which can function as a support and reference groupafter courses finish. One programme provided food rations, which were much appreciated: ‘Therewas a day I did not have anything to eat and SFP 10 –14 gave me a bag of food. The bag was hugeand it lasted for an entire month’ (Mejia et al., 2012, p. 62), and one provided cash transfers andsupport for accessing employment (Pereira Lima et al., 2007), though the evaluations did notattempt to measure the effects of these factors on learning or behaviour change. Integratedanti-poverty and parenting education programmes focused on younger children, such as ChileCrece Contigo (Daly et al., 2015) may have lessons for programmes focusing on the parents ofadolescents.

Expand discussion and deliberation around social normsOur analysis of programme curricula suggests that explicit discussion of norms is far more commonin programmes focusing on sexual violence and child marriage than in those focusing on physical

GLOBAL PUBLIC HEALTH 17

and emotional violence. The latter typically focus more on sharing information and practising com-munication skills13 (see Table 1). We know that norms condoning or even mandating violent disci-pline in certain situations are widespread, and that efforts to shift such norms among parents ofyoung children have achieved a considerable reduction in such violence; indeed, the connectionsbetween norms condoning intimate partner violence and child maltreatment mean that norms shift-ing programmes can be effective in reducing both simultaneously (Bacchus et al., 2017). Programmesaiming to reduce violence against adolescents would do well to learn from the structured and sys-tematic approach to changing harmful norms used by programmes such as REAL Fathers andSASA! in Uganda. These involved discussion of gendered social norms around violence and engagedparenting, as well as helping participants develop effective communication skills between couplesand with their children (Bacchus et al., 2017).

The way that programmes engage with norms is critical to their success. As compared to othernon-formal education programmes, parenting programmes more commonly draw on curriculadeveloped in other contexts, adapting them for local use (Marcus et al., 2019). This adaptationis critical for effectiveness and needs to encompass understanding of, and engagement with thespecifics of norms that influence violence against adolescents. Several evaluations (such as thoseof COMPASS in DRC and Parceria Project in Brazil) suggested that inadequate understandingof the local normative context affected participants’ receptiveness to programme content and con-tributed to the lack of change on a number of indicators. Indeed, Stark et al. (2018) suggest thatfurther efforts are needed to adapt parenting programmes to contexts of humanitarian crisis, givenspecific norms affecting parenting practices and perceptions of acceptable risks and levels of super-vision in these contexts. The Happy Families programme with Burmese refugees in Thailand rep-resents a promising example: the programme took as a point of departure the ‘meta-norm’ of‘loving kindness’ as a framework. This enabled facilitators to emphasise the ways suggested prac-tices were consistent with widely accepted norms and values, even where they required a behav-ioural shift and represented a challenge to connected norms, such as that children should obeyparents and that parents should take decisions without any need to consult their children (Simet al., 2014).

Invest in sustaining programme impactsThere is relatively little evidence of whether impacts have been sustained, one indicator of whethernorms are really changing. Only two studies (Paruk et al., 2009; Pereira Lima et al., 2007) reported onimpacts beyond two years. Promisingly, both found lasting impacts. However, two studies thatundertook surveys immediately post-participation and around six months later, found that behav-iour changes started to weaken over time (Cluver et al., 2018; Sim et al., 2014). Many of the strategiesdiscussed earlier (such as increasing systematic focus on norms, increasing engagement of fathersand norms influencers, strengthening peer support networks) are likely to contribute to a greaterdegree of sustainable change. In addition, programmes could consider holding refresher sessions,reunions, ad hoc follow-up support, or in contexts where access to mobile phones is common,reminders via text message or WhatsApp. These could all help sustain both in-person and virtualreference groups, reinforcing learning and behaviour change.

Conclusion

Overall, the evidence in our review suggests that parenting programmes appear to be contributing tothe building blocks of change in norms that underpin violence against adolescents, and if expanded,could extend impacts and catalyse a more substantial process of change. Given the huge scale of vio-lence against adolescents globally, and its long-lasting life consequences, our review suggests thatparent education programmes are an approach that deserves greater attention, as part of multi-faceted efforts to improve family wellbeing.

18 R. MARCUS ET AL.

Notes

1. Cure Violence Global. (Retrieved November 22, 2019).2. UNICEF Data: Child marriage around the world. (Retrieved December 16, 2019); UNICEF Data: Child mar-

riage among boys. (Retrieved December 16, 2019). Globally 21.2% of 20–24 year old women and 4.5% of20–24 year old men married in childhood.

3. For reasons of space, hereafter we will refer to parents and caregivers as ‘parents’ while recognising that not alladolescents live with their parents.

4. For two programmes, FMP and Sinovuyo Teen, it also draws on the available supplementary reports.5. Some of the studies reviewed additionally indicated that supplementary qualitative studies had taken place but

did not report their findings.6. Schwandt and Underwood (2013); Jejeebhoy et al. (2014).7. Sim et al. (2014); Jordans et al. (2013).8. Whether adolescents were included in any sessions was not clear for the other two programmes.9. Some studies only reported on study sample rather than the number of programme participants, thus possibly

biasing numbers downwards.10. In discussing these two forms of violence together, we are following the approach taken in majority of studies

reviewed.11. The two that did not were the SRH programme in Turkey (Kok & Akyüz, 2015) and COMPASS (Stark et al.,

2018).12. Families Matter! is currently active in 11 countries in Sub-Saharan Africa and Haiti. It is aimed at 9–12 year olds

and their parents with adolescents up to the age of 14 commonly participating (K. Miller, personal communi-cation, September 10, 2019).

13. It is important to note that our analysis may be affected by the level of description of curricula, which is highlyvariable between evaluations.

Acknowledgements

The authors wish to acknowledge funding from UK aid for the Gender and Adolescence: Global Evidence (GAGE)research initiative. For more information please see www.gage.odi.org

Disclosure statement

No potential conflict of interest was reported by the author(s).

References

ABAAD & Promundo. (2019). The pilot of program P-ECD in Lebanon: Evaluation results and lessons learned.Promundo. https://promundoglobal.org/wp-content/uploads/2020/01/PM001-P-ECD-Learning-Brief-OK.RR_.Final_.pdf

Annan, J., Sim, A., Puffer, E. S., Salhi, C., & Betancourt, T. S. (2017). Improving mental health outcomes of Burmesemigrant and displaced children in Thailand: A community-based randomized controlled trial of a parenting andfamily skills intervention. Prevention Science: The Official Journal of the Society for Prevention Research, 18(7),793–803. https://doi.org/10.1007/s11121-016-0728-2

Armistead, L., Cook, S., Skinner, D., Toefy, Y., Anthony, E. R., Zimmerman, L., Salama, C., Hipp, T., Goodnight, B., &Chow, L. (2014). Preliminary results from a family-based HIV prevention intervention for South African youth.Health Psychology, 33(7), 668–676. https://doi.org/10.1037/hea0000067

Bacchus, L. J., Colombini, M., Contreras Urbina, M., Howarth, E., Gardner, F., Annan, J., Ashburn, K., Madrid, B.,Levtov, R., & Watts, C. (2017). Exploring opportunities for coordinated responses to intimate partner violenceand child maltreatment in low and middle income countries: A scoping review. Psychology, health & medicine,22 (sup1). pp. 135–165. ISSN 1354-8506. doi:10.1080/13548506.2016.1274410.

Bacchus, L., Colombini, M., Contreras Urbina, M., Howarth, E., Gardner, F., Annan, J., & Watts, C. (2017). Exploringopportunities for coordinated responses to intimate partner violence and child maltreatment in low- and middle-income countries: A scoping review. Psychology, Health & Medicine, 22(sup1), 135–165. https://doi.org/10.1080/13548506.2016.1274410

Basu, S., Zuo, X., Lou, C., Acharya, R., & Lundgren, R. (2017). Learning to be gendered: Gender socialization in earlyadolescence among urban poor in Delhi, India, and Shanghai, China. Journal of Adolescent Health, 61(4S), S24–S29.https://doi.org/10.1016/j.jadohealth.2017.03.012

GLOBAL PUBLIC HEALTH 19

Bell, C., Bhana, A., Petersen, I., McKay, M., Gibbons, R., Bannon, W., & Amatya, A. (2008). Building protective factorsto offset sexually risky behaviors among black youths: A randomized control trial. Journal of the National MedicalAssociation, 100(8), 936–944. https://doi.org/10.1016/S0027-9684(15)31408-5

Bhana, A, Petersen, I, Mason, A, Mahintsho, Z, Bell, C, &Mckay, M. (2004). Children and youth at risk: adaptation andpilot study of the CHAMP (Amaqhawe) programme in South Africa. African Journal of AIDS Research, 3(1), 33–41.

Bogart, L. M., Skinner, D., Thurston, I. B., Toefy, Y., Klein, D. J., Hu, C. H., & Schuster, M. A. (2013). Let’s talk! A SouthAfrican worksite-based HIV prevention parenting program. Journal of Adolescent Health, 53(5), 602–608. https://doi.org/10.1016/j.jadohealth.2013.01.014

Buller, A. M., Pichon, M., McAlpine, A., Cislaghi, B., Heise, L., & Meiksin, R. (2020). Social norms linked to the sexualexploitation of children and adolescents: A global systematic review. Child Abuse and Neglect, 104. https://doi.org/10.1016/j.chiabu.2020.104471.

Carter, B., Apgar, M., & Khan Mohmand, S. (2019). Guidance note on scaling up social norm change. K4D EmergingIssues Report. Institute of Development Studies.

Cluver, L. D., Lachman, J. M., Ward, C. L., Gardner, F., Peterson, T., Hutchings, J. M., Mikton, C., Meinck, F.,Tsoanyane, S., Doubt, J., Boyes, M., & Redfern, A. (2017). Development of a parenting support program to preventabuse of adolescents in South Africa: Findings from a pilot pre-post study. Research on Social Work Practice, 27(7),758–766. https://doi.org/10.1177/1049731516628647

Cluver, L. D., Meinck, F., Steinert, J. I., Shenderovich, Y., Doubt, J., Herrero Romero, R., Lombard, C. J., Redfern, A.,Ward, C. L., Tsoanyane, S., Nzima, D., Sibanda, N., Wittesaele, C., De Stone, S., Boyes, M. E., Catanho, R., Lachman,J. M., Salah, N., Nocuza, M., & Gardner, F. (2018). Parenting for lifelong health: A pragmatic cluster randomisedcontrolled trial of a non-commercialised parenting programme for adolescents and their families in South Africa.BMJ Global Health, 3(1), e000539. https://doi.org/10.1136/bmjgh-2017-000539

Cluver, L., Meinck, F., Yakubovich, A., Doubt, J., Redfern, A., Ward, C., Salah, N., De Stone, S., Peterson, T.,Mpimpilashe, P., Romero, R. H., Ncobo, L., Lachman, J., Tsoanyane, S., Shenderovich, Y., Loening, H., Byrne, J.,Sherr, L., Kaplan, L., & Gardner, F. (2016). Reducing child abuse amongst adolescents in low- and middle-incomecountries: A pre-post trial in South Africa. BMC Public Health, 16(1), 567. https://doi.org/10.1186/s12889-016-3262-z

Corea, V. M. L., Zubarew, G. T., Valenzuela, M. M. T., & Salas, P. F. (2012). Evaluation of the program ‘strong families:Love and limits’ in families with teenagers aged 10 to 14 years. Revista Medica De Chile, 140(6), 726–731. https://doi.org/10.4067/S0034-98872012000600005

Cure Violence Global. Accessed December 22, 2019. https://cvg.org/Daly, M., Bray, R., Bruckauf, Z., Byrne, J., Margaria, A., Pećnik, N., & Samms- Vaughan, M. (2015). Family and par-

enting support: Policy and provision in a global context. Innocenti insights. UNICEF Office of Research Innocenti.Doubt, J., Bray, R., Loening-Voysey, H., Cluver, L., Byrne, J., Nzima, D., King, B., Shenderovich, Y., Steinert, J., &

Medley, S. (2017). ‘It has changed’: understanding change in a parenting program in South Africa. Annals ofGlobal Health, 83(5-6), 767–776. https://doi.org/10.1016/j.aogh.2017.10.021

Doubt, J., Loening-Voysey, H., Cluver, L., Byrne, J., Shenderovich, Y., Nzima, D., King, B., Medley, S., Steinert, J., &O’Malley, O. (2018). It empowers to attend: understanding how participants in the Eastern Cape of South Africaexperienced a parent support programme: A qualitative study. Innocenti Working Paper 2018-14. UNICEF Officeof Research.

Doyle, K., Levtov, R. G., Barker, G., Bastian, G. G., Bingenheimer, J. B., Kazimbaya, S., Nzabonimpa, A., Pulerwitz, J.,Sayinzoga, F., Sharma, V., & Shattuck, D. (2018). Gender-transformative Bandebereho couples’ intervention to pro-mote male engagement in reproductive and maternal health and violence prevention in Rwanda: Findings from arandomized controlled trial. PLoS ONE, 13(4), e0192756. https://doi.org/10.1371/journal.pone.0192756

Jejeebhoy, S. J., Francis Zavier, A. J., Santhya, K. G., Kumar Singh, S., Acharya, R., Gogoi, A., & Nathani, V. (2014).Promoting parent-child interaction and communication for healthy development of adolescents: Lessons from apilot project in rural Bihar. Population Council.

Jordans, M. J. D., Tol, W. A., Ndayisaba, A., & Komproe, I. H. (2013). A controlled evaluation of a brief parentingpsychoeducation intervention in Burundi. Social Psychiatry and Psychiatric Epidemiology, 48(11), 1851–1859.https://doi.org/10.1007/s00127-012-0630-6

Kamala, B. A., Rosecrans, K. D., Shoo, T. A., Al-Alawy, H. Z., Berrier, F., Bwogi, D. F., &Miller, K. S. (2017). Evaluationof the Families Matter! program in Tanzania: An intervention to promote effective parent-child communicationabout sex, sexuality and sexual risk reduction. AIDS Education and Prevention, 29(2), 105–120. https://doi.org/10.1521/aeap.2017.29.2.105

Klika, J. B., & Linkenbach, J. W. (2019). Social norms and violence against children and youth: Introduction to thespecial issue. Child and Adolescent Social Work Journal, 36, 1–3. https://doi.org/10.1007/s10560-018-0596-7

Knerr, W., Gardner, F., & Cluver, L. (2013). Improving positive parenting skills and reducing harsh and abusive par-enting in low- and middle-income countries: A systematic review. Prevention Science, 14(4), 352–363. https://doi.org/10.1007/s11121-012-0314-1

Know Violence in Childhood. (2017). Ending violence in childhood global report. New Delhi.

20 R. MARCUS ET AL.

Kok, G., & Akyüz, A. (2015). Evaluation of effectiveness of parent health education about the sexual developments ofadolescents with intellectual disabilities. Sexuality and Disability, 33(2), 157–174. https://doi.org/10.1007/s11195-015-9400-1

Learning Collaborative. (2017). Identifying and describing approaches and attributes of norms-shifting interventions.Institute for Reproductive Health, Georgetown University.

Learning Collaborative. (2019). Considerations for scaling up norms-shifting interventions for adolescent and youth sex-ual and reproductive health. Institute for Reproductive Health, Georgetown University.

Lilleston, P. S., Goldmann, L., Verma, R. K., & McCleary-Sills, J. (2017, March). Understanding social norms and vio-lence in childhood: Theoretical underpinnings and strategies for intervention. Psychology, Health & Medicine, 22(Suppl 1), 122–134. https://doi.org/10.1080/13548506.2016.1271954

Loening-Voysey, H., Doubt, J., Nzima, D., Shenderovich, Y., Steinert, J., Byrne, J., & Cluver, L. (2018). Relevance,implementation and impact of the Sinovuyo Teen Parenting Programme in South Africa. Innocenti ResearchReport. UNICEF Office of Research – Innocenti.

Lundgren, R., Gibbs, S., & Kerner, B. (2018). Does it take a village? Fostering gender equity among early adolescents inNepal. International Journal of Adolescent Medicine and Health, Advance online publication. https://doi.org/10.1515/ijamh-2017-0164

Maalouf, W., & Campello, G. (2014). The influence of family skills programmes on violence indicators: Experiencefrom a multi-site project of the United Nations office on Drugs and Crime in low- and middle-income countries.Aggression and Violent Behavior, 19(6, SI), 616–624. https://doi.org/10.1016/j.avb.2014.09.012

Marcus, R., Kruja, K., & Rivett, J. (2019). What are the impacts of parenting programmes on adolescents? GAGE,Overseas Development Institute. https://www.gage.odi.org/wp-content/uploads/2020/01/Parenting-Prog-Review-v04.pdf.

Mejia, A., Calam, R., & Sanders, M. R. (2012). A review of parenting programs in developing countries: Opportunitiesand challenges for preventing emotional and behavioral difficulties in children. Clinical Child and FamilyPsychology Review, 15(2), 163–175. https://doi.org/10.1007/s10567-012-0116-9

Miller, K. S., Winskell, K., & Berrier, F. L. (2016). Responding to changes in HIV policy: Updating and enhancing theFamilies Matter! Curriculum. Health Education Journal, 75(4), 409–420. https://doi.org/10.1177/0017896915595530

Orpinas, P., Ambrose, A., Maddaleno, M., Vulanovic, L., Mejia, M., Butron, B., Gutierrez, G., & Soriano, I. (2014).Lessons learned in evaluating the Familias Fuertes program in three countries in Latin America. RevistaPanamericana de Salud Publica [Pan American Journal of Public Health], 36(6), 383–390.

Pan American Health Organisation. (2006). Informe de la Fase II del Programa Piloto ‘Familias Fuertes. Amor yLímites’. Phase II Report of the Pilot Programme ‘Strong Families. Love and Limits’.

Panter-Brick, C., Burgess, A., Eggerman, M., McAllister, F., Pruett, K., & Leckman, J. F. (2014). Practitioner review:Engaging fathers – recommendations for a game change in parenting interventions based on a systematic reviewof the global evidence. Journal of Child Psychology and Psychiatry, 55(11), 1187–1212. https://doi.org/10.1111/jcpp.12280

Paruk, Z., Petersen, I., & Bhana, A. (2009). Facilitating health-enabling social contexts for youth: Qualitative evaluationof a family-based HIV-prevention pilot programme. African Journal of AIDS Research, 8(1), 61–68. https://doi.org/10.2989/AJAR.2009.8.1.7.720

Pereira, P., D’Affonseca, S., & Williams, L. (2013). A feasibility pilot intervention program to teach parenting skills tomothers of poly-victimized children. Journal of Family Violence, 28(1), 5–15. https://doi.org/10.1007/s10896-012-9490-9

Pereira Lima, V. L., Arruda, J. M., Barroso, M. A., Lobato Tavares, M. F., Ribeiro Campos, N. Z., Zandonadil, R. C., daRocha, R. M., Parreira, C. M., Cohen, S. C., Kligerman, D. C., Sperandio, A. M., Correa, C. R., & Serrano, M. M.(2007). Analyzing the outcomes of health promotion practices. Promotion & Education Supplement, 1, 21–26.https://doi.org/10.1177/10253823070140010501x

Puffer, E. S., Annan, J., Sim, A. L., Salhi, C., & Betancourt, T. S. (2017). The impact of a family skills training inter-vention among Burmese migrant families in Thailand: A randomized controlled trial. PLoS One, 12(3),e0172611. https://doi.org/10.1371/journal.pone.0172611

Puffer, E. S., Green, E. P., Sikkema, K. J., Broverman, S. A., Ogwang-Odhiambo, R. A., & Pian, J. (2016). A church-based intervention for families to promote mental health and prevent HIV among adolescents in rural Kenya:Results of a randomized trial. Journal of Consulting and Clinical Psychology, 84(6), 511–525. https://doi.org/10.1037/ccp0000076

Schwandt, H. M., & Underwood, C. (2013). Making a difference in adult-child relationships: Evidence from an adult-child communication intervention in Botswana, Malawi, and Mozambique. Journal of Adolescence, 36(6), 1177–1186. https://doi.org/10.1016/j.adolescence.2013.09.005

Shaw, S., Cham, H., Lin, C., Galloway, E., Winskell, K., Kasese, C., Bangani, Z., & Miller, K. S. (2019). Engaging parentsand caregivers to prevent and respond to child sexual abuse in sub-Saharan Africa: Results from an intervention studyin Zimbabwe.

GLOBAL PUBLIC HEALTH 21

Sim, A., Annan, J., Puffer, E., Salhi, C., & Betancourt, T. (2014). Building happy families: Impact evaluation of a par-enting and family skills intervention for migrant and displaced Burmese families in Thailand. International RescueCommittee.

Stark, L., Seff, I., Asghar, K., Roth, D., Bakamore, T., MacRae, M., Fanton D’Andon, C., & Falb, K. L. (2018). Buildingcaregivers’ emotional, parental, and social support skills to prevent violence against adolescent girls: Findings from acluster randomised controlled trial in Democratic Republic of Congo. BMJ Global Health, 3(5), e000824. https://doi.org/10.1136/bmjgh-2018-000824

UNICEF Data. Child marriage among boys. Retrieved December 16, 2019. https://data.unicef.org/resources/child-marriage-among-boys-available-data/.

UNICEF Data. Child marriage around the world. Retrieved December 16, 2019. https://www.unicef.org/stories/child-marriage-around-world.

U.S. Centers for Disease Control and Prevention (CDC). (2019). Families matter! Program overview [Unpublishedmimeo].

Valente, J. Y., Moreira, T. C., Ferigolo, M., & Barros, H. M. T. (2018). Randomized clinical trial to change parentalpractices for drug use in a telehealth prevention program: A pilot study. Jornal de Pediatria, 95(3), 334–341.https://doi.org/10.1016/j.jped.2018.02.004

Vandenhoudt, H., Miller, K. S., Ochura, J., Wyckoff, S. C., Obong’o, C. O., Otwoma, N. J., Poulsen, M. N., Menten, J.,Marum, E., & Buve, A. (2010). Evaluation of a U.S. Evidence-based parenting intervention in rural western Kenya:From parents matter! To families matter!. AIDS Education and Prevention, 22(4), 328–343. https://doi.org/10.1521/aeap.2010.22.4.328

Vasquez, M., Meza, L., Almandarez, O., Santos, A., Matute, R. C., Canaca, L. D., Cruz, A., Acosta, S., Garcia Bacilla, M.E., Wilson, L., Azuero, A., Tami, I., Holcomb, L., & Saenz, K. (2010). Evaluation of a strengthening families(Familias Fuertes) intervention for parents and adolescents in Honduras. Southern Online Journal of NursingResearch, 10(3), 13.

Wessels, I., Mikton, C., Ward, C. L., Kilbane, T., Alves, R., Campello, G., Dubowitz, H., Hutchings, J., Jones, L., Lynch,M., & Madrid, B. (2013). Preventing violence: Evaluating outcomes of parenting programmes. World HealthOrganization.

World Health Organisation. (2018). INSPIRE handbook: Action for implementing the seven strategies for ending vio-lence against children. Geneva: World Health Organisation. ISBN: 9789241514095.

22 R. MARCUS ET AL.